Systems and methods for preparing compositions

Abstract

In one embodiment, a method includes receiving identities of features of a first composition, and receiving a value of a property of the first composition. A learning model is trained using values of predefined characteristics of the features of the first composition, and the value of the property of the first composition. The learning model is then provided identities of second features of a second composition. Using the learning model, a value of a property of the second composition is determined and displayed.

What is claimed is: 
     
       1. A method comprising:
 (a) receiving identities of features of a first composition, each of the features having a value of a predefined characteristic of predefined characteristics; 
 (b) receiving a value of a property of the first composition, wherein the property is affected by an interaction of at least two of the features of the first composition; 
 (c) training a learning model using the values of the predefined characteristics of the features of the first composition and the value of the property of the first composition; 
 (d) providing to the learning model identities of second features of a second composition, wherein at least one of the second features is different than at least one of the first features; 
 (e) determining, via the learning model, a value of a property of the second composition; and 
 (f) causing a user interface to display an indication of the value of the property of the second composition for producing a product comprising the second composition; 
 wherein steps (c)-(e) are performed by one or more processors. 
 
     
     
       2. The method according to  claim 1  wherein the property of the second composition is affected by an interaction of at least two of the features of the second composition. 
     
     
       3. The method according to  claim 2  wherein at least one of the features of the first composition is an ingredient, and at least one of the features of the second composition is an ingredient. 
     
     
       4. The method according to  claim 1  wherein the property of the first composition and the property of the second composition are the same property. 
     
     
       5. The method according to  claim 1  wherein the property of the first composition and the property of the second composition are different properties, the property of the second composition being derivable from the property of the first composition. 
     
     
       6. The method according to  claim 1  wherein the learning model is a machine learning model, the property of the second composition being determined via rules of the machine learning model. 
     
     
       7. The method according to  claim 6  wherein the machine learning model comprises a supervised learning approach. 
     
     
       8. The method according to  claim 7  wherein the supervised learning approach comprises at least one of a Decision Tree, Random Forest, Support Vector Machine, Naïve Bayes Classification, or Logistic Regression rule. 
     
     
       9. The method according to  claim 1  wherein the property of the second composition relates to a physiochemical property. 
     
     
       10. The method according to  claim 1  wherein the identities of the features of the first composition and the identities of the features of the second composition are stored in a database. 
     
     
       11. The method according to  claim 1  wherein the value of the property of the first composition is identified via at least one of an experimental measurement of the first composition or a thermodynamic calculation of the first composition. 
     
     
       12. The method according to  claim 1  wherein the value of the property of the first composition is identified via an experimental measurement of the first composition. 
     
     
       13. The method according to  claim 1  wherein the value of the property of the first composition is comprised of at least one of a first value identified via an experimental measurement of the sample composition or a second value identified via a thermodynamic calculation. 
     
     
       14. The method according to  claim 1  wherein the second composition forms part of a personal care product selected from: a toothpaste; a mouthwash; a shower gel; a body wash; a liquid hand soap; a fabric softener; a shampoo; a conditioner; a lotion; a cream; and an ointment. 
     
     
       15. The method according to  claim 1  wherein each of the second features of the second composition has a predefined characteristic having a value. 
     
     
       16. The method according to  claim 15  wherein the predefined characteristic is a chemoinformatic property. 
     
     
       17. The method according to  claim 1  comprising:
 repeating steps (a)-(c) for a plurality of first compositions prior to performing step (d)-(e). 
 
     
     
       18. A method comprising:
 (a) receiving values of chemoinformatic properties of ingredients of a sample chemical composition; 
 (b) receiving a value of a property of the sample chemical composition, the property being affected by an interaction of at least two of the ingredients; 
 (c) inputting, into a model, the values of the chemoinformatic properties of the ingredients of the sample chemical composition and the value of the property of the sample chemical composition; 
 (d) determining, via the model, an identity of a second chemical composition based on at least one of (1) values of chemoinformatic properties of ingredients of the second chemical composition or (2) a value of a property of the second chemical composition; and 
 (e) causing a user interface to display an indication of the identity of the second chemical composition for producing a product comprising the second composition; 
 wherein steps (c)-(d) are performed by one or more processors. 
 
     
     
       19. The method of  claim 18  wherein the determining of the identity of the second chemical composition is based on the values of the chemoinformatic properties of the ingredients of the second chemical composition and not based on the value of the property of the second chemical composition. 
     
     
       20. The method of  claim 18  wherein, for each of the sample chemical composition and the second chemical composition, each of the ingredients has a respective chemoinformatic property.